101. Respondents to the surveys have noted that the current IAS 36 in some cases provides little or no specific guidance. Academic research (e.g. Glaum et al., 2013; Peterson et al., 2010) confirms that there is uncertainty as to how IAS 36’s requirements are applied in practice.
102. In particular, respondents have identified the following areas of judgment:
(a) the choice of CGUs to which goodwill is allocated; and
Should Goodwill still not be Amortised?
Accounting and Disclosure for Goodwill
36
Identification of CGUs to which goodwill is allocated
103. In relation to the identification of CGUs, IAS 36 paragraph 68 says that: “an asset’s cash- generating unit is the smallest group of assets that includes the asset and generates cash inflows that are largely independent of the cash inflows from other assets or groups of assets. Identification of an asset’s cash-generating unit involves judgment. If recoverable amounts cannot be determined for an individual asset, an entity identifies the lowest aggregation of assets that generate largely independent cash inflows.”
104. Size and composition of CGUs impacts the recognition and measurement of impairment losses, but IAS 36 offers limited guidance in identifying them. If a CGU is too broad, it may result in non-recognition of impairment because the gains of some units may offset the losses of another unit within the same CGU. Paragraph 80b of IAS 36 requires that a CGU should not be larger than an operating segment determined in accordance with IFRS 8 Segment Reporting, but some studies cast doubts on whether there are entities that do not comply with this requirement.
105. Based on empirical research, Ernst & Young (2006, 35) concludes that frequently entities identify only one CGU for each of their segments, even if they do not specifically disclose this fact. Contrary to that, Petersen and Plenborg (2010) on the basis of a survey conducted across all Danish listed companies, find that only 25% of the entities surveyed identify CGUs at the segment level. This may indicate that in some jurisdictions preparers are more rigorous in applying the requirements.
106. In relation to the allocation of goodwill to the CGUs, it should be noted that a different approach exists between IFRSs and US GAAP. Under US GAAP, goodwill is allocated to the reporting unit level whilst under IFRS it is allocated to the CGUs. Reporting unit is defined in US GAAP as an operating segment or one level below an operating segment (component).
107. The Research Group believes that to favour a consistent application of the standard a stricter guidance could be introduced to facilitate the allocation of goodwill to the smallest CGU.
Inputs and methods for estimating the recoverable amount of CGUs
108. In relation to the inputs and methods for estimating the recoverable amount, the Research Group believes that the areas in need of improvement are:
(a) the valuation approach;
(b) the calculation of the VIU;
(c) the discount rates and cash flows; and
Should Goodwill still not be Amortised?
Accounting and Disclosure for Goodwill
37 109. As mentioned above, paragraph 18 of IAS 36 states that the recoverable amount is the
higher of its FVLCTS and its VIU. On this point, the OIC-EFRAG survey shows that entities mostly use VIU in determining the recoverable amount. In all likelihood, the infrequent use of FVLCTS is due to the fact that it requires information on the market value of the unit.
110. Some have argued that the method applied to determine the recoverable amount should reflect the expected manner of recovery, and if the entity intends to recover the value by operating the unit, the recoverable amount should only be based on VIU.
111. To estimate VIU, paragraph 30 of IAS 36 requires applying the discounted cash flow model (DCF), including its variants such as the dividend discount model (DDM). However, there are studies that show that some entities apply other methods such as EVA-model (economic value added model) and multiples (Peterson et al. 2010).
112. IFRS 13 has clarified methods for the calculation of fair value in the absence of a primary market for the asset being valued. It has also explained that fair value of a non- financial asset should represent its highest and best use. Some question that in very few circumstances an entity could conclude that value in use may be higher than a fair value if the latter represents its highest and best use. Such an aspect may be considered by the IASB in revisiting the requirement of IAS 36.
Assessing the discount rate
113. A significant area of judgement in the calculation of the VIU is the determination of the discount rate. Paragraph 55 of IAS 36 requires that an entity must use a pre-tax discount rate to discount pre-tax cash flows so there is consistency between the discount rate and the cash flows. Paragraph 94 of the Basis for Conclusions of IAS 36 states that, conceptually discounting post-tax cash flows at a post-tax discount rate or discounting pre-tax cash flows at a pre-tax discount rate should give the same results. This could be understood as an indirect equivalence between pre-tax and post-tax discount rates. However, if the expected future cash flows are not evenly distributed over the period, the equivalence may not occur.
114. In addition, most academic books (e.g. Koller et al., 2005) estimate value in use on the basis of post-tax cash flows, discounted with a post-tax discount rate.
115. IAS 36 is not explicit as to how to estimate a pre-tax discount rate, and implicitly suggests using an iterative process in certain cases (see paragraphs 56, 57 and BCZ85). It would seem that VIU could be estimated starting from a post-tax calculation to arrive at an equivalent result through an iterative process. However, as said above, the iterative process depends upon the time distribution of the cash flows. Moreover, entities can be consistent with the concept stated in paragraph 94 of the Basis for Conclusions in IAS 36 only if the cash flows are uniformly distributed along the time period. More particularly, if the cash flows are variable and/or uneven, the discount factor will influence the results according to how the flows are distributed.
Should Goodwill still not be Amortised?
Accounting and Disclosure for Goodwill
38
116. The examples and explanations in IAS 36 do not clearly explain how the iterative process should be applied when the cash flows are uneven and the entity is unable to reconcile between the post and pre-tax rate and flow. The reconciliation is possible when the flows and the growth rate are even, without the need for an iterative process but simply by grossing-up the post-tax rate and flow. The IASB should then clarify that using the formula [post-tax rate/(1 – corporate tax rate] to determine the pre-tax rate works only when no growth is assumed in the future periods. If the growth rate is positive, and the flows are assumed to be steady over the period, it is still possible to determine the pre- tax rate by grossing up both the post-tax rate and the growth rate.
117. The issue related to the calculation of the discount rate is not limited to the tax effect but also whether there is a consistency among entities using IFRSs in relation to the input used in determining the discount rate. To this end the Research Group looks favourably to the research work initiated by the IASB staff on the use of discount rates and is keen to be involved in the project.
118. Paragraph 33b of IAS 36 requires that projections based on budgets and forecasts shall cover a maximum period of five years, unless a longer period can be justified. The IASB intended limiting the explicit forecast period because the entity’s ability to forecast beyond 5 years is presumed to be low. Yet IAS 36 does not provide any guidance on the methods to estimate the terminal value (e.g. Gordon’s growth model, a value driver formula, or others). Studies show that some entities use multiples to estimate terminal value: Petersen and Plenborg examined a sample of Danish companies and found that some use methods other than a DCF model such as EVA (Evaluation Value Added model), multiples and other methods.
119. More specifically, these studies concluded that some preparers use a DCF model only for the explicit period and then add a terminal value calculated under a different method. Although IAS 36 does not prohibit this, it might be inconsistent with a DCF model. In all likelihood, it can be assumed that considering an infinite lifetime the majority of the recoverable amount can be attributed to the terminal value, and in such circumstances most of the value estimated is captured by the multiple.
Should Goodwill still not be Amortised?
Accounting and Disclosure for Goodwill
39
CONCLUSION
120. The Research Group thinks that the following conclusion can be drawn from the above analysis:
(a) developing more prescriptive requirements and guidance to reduce the recognition of internally generated goodwill is impracticable. Such an issue can be addressed in part by improving the impairment test model but only if the improvement is combined with the re-introduction of the amortisation of goodwill. In any case, the impairment test needs to be retained even if amortisation of goodwill is reintroduced;
(b) there are a number of areas for possible improvements of IAS 36 in order to reduce the operational challenges. This applies to calculation of VIU and its relationship with fair value as well as the determination of the discount rate; and
(c) in relation to the concept and usage of discount rate, the Research Group is interested in the research work conducted by the IASB staff and is keen to be involved in the project.
Should Goodwill still not be Amortised?
Accounting and Disclosure for Goodwill
40